Module COMP3741: PARALLEL SCIENTIFIC COMPUTING

Department: Computer Science

COMP3741: PARALLEL SCIENTIFIC COMPUTING

Prerequisites

• COMP1071 Computer Systems AND COMP2221 Programming Paradigms AND (COMP1021 Maths for Computer Science OR MATH1551 Maths for Engineers and Scientists OR (MATH1561 Single Mathematics A AND MATH1571 Single Mathematics B) OR (MATH1061 Calculus I AND MATH1017 Linear Algebra I))

Corequisites

• None

Excluded Combination of Modules

• MATH3081 Numerical Differential Equations III AND MATH4221 Numerical Differential Equations IV

Aims

• To develop an understanding of current hardware architectures and how they affect programming models, application performance and security 
• Introduce and familiarise students with parallel programming approaches 
• Introduce scientific computing techniques for the numerical solution of problems in science and engineering
• To develop practical skills in application development, server provisioning and orchestration in the context of large scale computing

Content

• Characteristics and design of current processor architectures 
• Memory systems: cache design and performance 
• Processor architecture and security 
• Data parallel programming for shared memory, distributed memory and heterogeneous systems 
• Fundamentals of algorithmic accuracy and performance engineering for applications in scientific computing
• Fundamentals of numerical algorithms for ordinary differential equations 
• System administration

Learning Outcomes

• On completion of the module, students will be able to demonstrate: 
• an understanding of typical approaches to the numerical solution of problems in science and engineering. 
• knowledge and appreciation of some of the research challenges in scientific computing 
• an understanding of basic parallelisation strategies and when to apply them 
• the relationship between hardware architectures and high level programming languages 
• complex performance issues of current computers, and the selection of processors for specific applications

• On completion of the module, students will be able to demonstrate: 
• an awareness of current technology, design analysis, and commercial practice and the ability to bring these together to provide innovative solutions for digital systems 
• an ability to apply numerical techniques to solve ordinary differential equations 
• an ability to develop appropriate parallelisation schemes 
• an ability to critically evaluate how the subject knowledge could be applied to various applications

• On completion of the module, students will be able to demonstrate: 
• an ability to propose appropriate solutions to problems in scientific computing 
• an ability to communicate technical information
• capacity for self-learning in familiar and unfamiliar situations

Modes of Teaching, Learning and Assessment and how these contribute to the learning outcomes of the module

• Lectures provide the students with the theoretical background of the material, and aid understanding with examples. 
• Practicals enable students to acquire the necessary coding skills, learn about the relevant programming models, libraries and packages and receive feedback on their work. 
• Summative assessments assess the knowledge of relevant libraries and application of methods and techniques.

Teaching Methods and Learning Hours

Summative Assessment

Formative Assessment:

Example formative exercises are given during the course. The lab practicals contain formative assignments. Feedback will be provided to the students on the summative assignments and lecture materials during the practicals.

■ Attendance at all activities marked with this symbol will be monitored. Students who fail to attend these activities, or to complete the summative or formative assessment specified above, will be subject to the procedures defined in the University's General Regulation V, and may be required to leave the University